From: | Mike Mascari <mascarm(at)mascari(dot)com> |
---|---|
To: | Tim Perdue <tperdue(at)valinux(dot)com> |
Cc: | pgsql-hackers(at)hub(dot)org |
Subject: | Re: 7.0.2 issues / Geocrawler |
Date: | 2000-07-12 11:34:13 |
Message-ID: | 396C57B5.A59B0F63@mascari.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Tim Perdue wrote:
>
> This is a *big* help.
>
> Yes, the table is approx 10-12GB in size and running your length() and
> update queries is going to take a lifetime, since it will require a
> calculation on 4 million rows.
>
> This doesn't address the serious performance problem I'm finding in
> 7.0.2 for a multi-key select/order by/limit/offset query, which I sent
> in a separate email.
>
> Tim
If I recall correctly, Marc experienced similar performance
differences with UDM search after upgrading. The optimizer was
redesigned to be smarter about using indexes with both order by
and limit. Tom Lane, of course, knows all there is to know on
this. All I can ask is standard issue precursor to optimizer
questions:
Have you VACUUM ANALYZE'd the table(s) in question?
If so, hopefully Tom Lane can comment.
Sorry I couldn't be more help,
Mike Mascari
From | Date | Subject | |
---|---|---|---|
Next Message | Peter Mount | 2000-07-12 11:36:02 | RE: Contacting me |
Previous Message | Tim Perdue | 2000-07-12 11:27:51 | Re: 7.0.2 issues / Geocrawler |